Engineering With Large Language Models

Welcome to my page, EngineeringwithLLMs.info. My name is William Katzenmeyer, P.E., C.F.M. and I am a Senior Water Resources Engineer with over 16 years of experience across multiple disciplines including hydraulic and hydrologic modeling and automation development, flood risk management, FEMA grant management (HMGP/PA/BRIC), Benfit Cost Analysis (BCA) technical assistance, snd civil site and utility design. My passion lies at the intersection of civil engineering and the frontier of artificial intelligence: Large Language Models (LLM's). I've pioneered and shared innovative approaches to streamline and enhance traditional water resources workflows, as well as sharing my methods in an open-source manner to foster exploration and responsible adoption of large langauge models across the civil engineering profession.

While facing my own challenges of tackling complex hydraulic modeling projects that demanded intense computational resources, Large Language Models (LLMs) burst into existence and enabled me to tackle increasingly complex scripting tasks. This ultimately led me to create and share the HEC-Commander tools repository: an open-source collection of Python notebooks developed with AI assistance that dramatically improves efficiency in HEC-RAS and HEC-HMS calibration and validation workflows, blogs and ChatGPT examples, as well as providing the only open source method for paralell execution of HEC-RAS instances across multiple networked windows machines, directly from a jupyter notebook without containerization or use of expensive cloud compute platforms.

My work represents a new paradigm in civil engineering: leveraging Large Language Models (LLMs) to provide targeted assistance, help with research and learning, and automating repetitive tasks and unlock new capabilities for practicing engineers to creatively solve complex problems. The thoughtful application of LLM capabilities combined with traditional engineering expertise holds immense potential, especially where ideas can be quickly and transparently implemented to assist practicing engineers with completing their everyday tasks as well as exapanding the frontier of what they can achieve by allowing the innvoation of new capabilities through code.

As an open source developer, educator and speaker, I'm committed to sharing these innovations with the broader engineering community. Through presentations, workshops, and instructional content, I have helped other engineers adopt AI-driven approaches that can transform their own workflows and projects. I look forward to contributing to the ongoing evolution of how civil engineers leverage technology to enhance their work. See my "Upcoming Appearances" section below to see where to find me next!

RAS-Commander is a comprehensive Python library for automating HEC-RAS operations, providing a modern alternative to the HECRASController API. Building on the foundation of my HEC-Commander tools, this library offers powerful capabilities for water resources engineers, including project management, HDF data access, and ASCII file operations.

RAS-Commander is designed from the ground up to be used in conjunction with Large Language Models for scripting support. The library features purpose-built knowledge bases, Cursor IDE instructions, a specialized library assistant and a ChatGPT assistant. Whether you're managing multiple model scenarios, extracting detailed results data, or implementing automated QA/QC processes, this library provides the tools you need to work more efficiently with HEC-RAS.

HEC-Commander Tools is a suite of Python notebooks developed with AI assistance for water resource engineering workflows, primarily focused on providing automation for HEC-RAS and HEC-HMS through Jupyter Notebooks. This open-source project includes tools, blogs, and ChatGPT assistants relevant to H&H modeling, automation, and the use of LLMs for water resources workflows.